Contact us
Product Huntlaunching December 5 Follow the launch

Research · 7 min read

Why AI product videos get the size wrong, and how to check it

AI generators don't know how big your product is, so they guess from the product type. In our test, four image generators drew a 41 cm massage roller at a median 1.47 times its real size, while a familiar tote bag came out within 10%. Writing the real size into the prompt barely helped.

A black massage roller on a dark plinth, with a larger translucent copy of it behind, the size an AI generator drew it at

Key takeaways

  • In our November 2026 test, four image generators drew a 41 cm massage roller at a median 1.47× its real size next to a 168 cm woman. None of 32 frames came within 10% of true size.
  • A familiar product came out right: a tote bag was drawn at a median 1.03×, with 24 of 27 frames within 10%. Generators know common products and guess at the rest.
  • Real dimensions in the prompt moved the roller only from 1.48× to 1.45× on median. A prompt describes the size; nothing in the generator checks it.
  • Image-to-video models animate the first frame they are given, so a size error in that frame carries into the whole clip, and every new shot can add another one.
  • What works is measurement: build the product at its real dimensions first, check every frame against it and regenerate the frames that drift.

AI video for product pages has gone mainstream. Alibaba says its content tools for merchants generate about 5 million videos a month, and its video feature works from just a few product photos. TikTok Shop’s seller center has a tool that turns product images into shoppable videos. Before a catalog team puts such a video next to the Add to cart button, one question is worth a number: is the product the right size?

We tested it.

How far off is the size in AI product images?

Far enough to see without a ruler. We gave four widely used image generators the same job: a woman stands full-length and holds a product. Two products, two prompts, four takes each: 64 frames. Then we measured the product against the woman in every frame.

Eight generated frames: four women holding a massage roller drawn 1.43 to 1.51 times its real length, and four women holding a padel tote bag drawn 1.01 to 1.05 times its real size, each with a red line for the drawn length and a jade line for the true length
One frame per generator, the take closest to its median. Red: the length the generator drew. Jade: the true length at the same place, worked out from the woman's height. Top row, the roller: 1.43× to 1.51×. Bottom row, the tote bag: 1.01× to 1.05×.

The roller is a powered massage roller from a fitness brand we made test videos for. It is 40.8 cm from one handle tip to the other, so next to a 168 cm woman it should be a little wider than her shoulders. The generators drew it at a median 1.47 times that length, anywhere from 1.15× to 1.72×. Not one of 32 frames came within 10% of the real size.

The tote bag behaved differently. Its body is 33.2 cm tall, and the generators drew it at a median 1.03×. 24 of 27 frames landed within 10%.

Data table
Product · promptFramesMedian sizeRangeWithin 10%
Massage roller, 41 cm, no sizes in the prompt16×1.48×1.39–×1.720 of 16
Massage roller, 41 cm, real sizes in the prompt16×1.45×1.15–×1.660 of 16
Padel tote bag, 33 cm tall, no sizes in the prompt13×1.04×0.93–×1.1111 of 13
Padel tote bag, 33 cm tall, real sizes in the prompt14×1.01×0.96–×1.1113 of 14
Every dot is one generated frame, measured against a 168 cm woman in the same frame. The roller came out at a median ×1.47 across 32 frames, and none landed within 10% of its true size; the tote bag came out at a median ×1.03, with 24 of 27 frames within 10%. Quelta test, November 5, 2026: four image generators, four takes per prompt.

Why does AI get some products right and others wrong?

A generator learns what things look like, and how big they are, from the photos it was trained on. Tote bags appear next to people in a huge number of photos, so the model has a firm sense of their size. A powered massage roller with two handles is far rarer. The model has to infer its size from what the product resembles.

The reference photo doesn’t help. A packshot on a white background has no scale: nothing in it says whether the roller is 40 cm or 70 cm long. So the generator falls back on the closest product type it knows. Our roller has a ribbed drum and two straight handles held with both hands, which puts it close to gym equipment. That part is our reading of the results. The test shows that the size is wrong, not why.

What the test does show is that the error is systematic. All four generators missed in the same direction and by a similar amount: their medians for the roller were 1.44×, 1.44×, 1.49× and 1.52×. When four different systems agree on the wrong answer, generating one more take won’t fix it.

For a catalog, this is the uncomfortable part. Generators do well on common products, and common products are where AI video adds little. New, technical or unusual products are the ones that need video, and a generator knows little about them.

Does writing the dimensions in the prompt help?

Barely. With the real sizes in the prompt (“the roller is 41 cm long… the woman is 168 cm tall. Keep these real proportions exactly”), the roller’s median moved from 1.48× to 1.45×. The closest single frame reached 1.15×, but 12 of the 16 frames with sizes in the prompt were still more than 30% too big.

We had seen the same in production before this test. In one frame, the prompt gave a woman’s height in centimeters, and the generator still drew her about 1.2 times too tall for the sofa next to her.

Generated frame: a woman with a tray beside a modular sofa, drawn about 1.2 times too tall; the dashed outline is a true-size figureGenerator
Quelta frame: the woman by the ottoman fills the dashed outline of a true-size figureQuelta
Left: the prompt gave her height, yet she came out about 1.2× too tall for the sofa. Right: our frame, where she matches a true-size 3D figure.

A prompt describes the size. Nothing in the generator checks it.

Where else does size break in a product video?

The first frame is only where the problems start. Most AI product videos are made in two steps: a generator draws a still frame with the product, then an image-to-video model animates it. The video model takes that first frame as given, so a size error in it runs through the whole clip. After that, three more kinds of errors appear:

  • Between shots. Every shot is drawn again from scratch, so a sofa can grow and shrink from one shot to the next.
  • Between angles. Seen from another side, the product no longer matches itself. In one of our tests, two shots of the same bed in a row disagreed: in the second, the rail post was 1.8 times taller and the frame sat 20% lower. Each frame looked fine on its own.
  • Inside a shot. When a hand picks the product up or the camera moves, the product can stretch, lose a part or change its proportions between the first and the last frame.
First shot of an opened sofa bed with reference lines at the frame height, along the leg and around the rail postShot 1
Next shot of the same bed: the frame sits lower, the leg leans and the rail post is tallerShot 2
Two generated shots of the same bed, back to back. In the second, the rail post is 1.8× taller and the frame sits 20% lower.

This matters because video on a product page is there to set expectations. Amazon tells sellers that video on product pages can decrease returns by more than 40%, based on its internal data for 2022. That only works if the video shows the product as it is. In a survey by Salsify of 1,910 U.S. and U.K. shoppers, 71% said they had returned an item because of incorrect product content.

How to check an AI product video before it goes live

Measure, don’t eyeball. A product that is 20% too big looks normal on its own; you only see it next to something whose size you know. Pick a frame, find a reference in it, such as a person, a hand, a door or a table, and measure both in pixels. Any image viewer shows pixel coordinates.

Real life, cm
In the frame, pixels
×1.47 The product is drawn 47% too big. Regenerate the shot.
Size factor = (product ÷ reference in the frame) ÷ (product ÷ reference in real life). Measure both lines in the same frame and at about the same distance from the camera. We treat ±5% as a pass, 5–10% as worth a second look, and anything beyond 10% as a reshoot. The defaults are a frame from our test.

Size is the first check. These are all eight we run on every frame before a video leaves our production. The file itself, its length, format and resolution for each marketplace, is a separate check: our free video requirements checker runs it in the browser.

Tick each check the video passes.
The checks we run on every frame before a video leaves our production. Tick the ones your video passes.

If a vendor’s video fails the first two checks, ask how they set the size, not for a revision. A revision is another guess.

How we keep the size right in production

This is the problem our engine was built around. Before any frame is generated, it builds a simple 3D block-out of the product at its real dimensions, from the product photos and the sizes on the product page. The block-out fixes the size and proportions; the look of the product comes from the photos. Every frame is generated around the product’s outline and then checked against the model and the photos: size, shape, number of parts, logo and color. A frame that fails is regenerated.

The check is what makes it hold. For one product video for a fitness brand, we generated 53 candidate frames for 15 shots with the 3D guide. Only 27 of them, about half, stayed within 6 pixels of the guide’s outline at a frame width of 1,080 pixels. Keeping the closest take for each shot put 14 of 15 shots within 6 pixels, with a median of 3. Without the check, half of the frames would have reached the video with the product off.

The padel bag from our size test, in a video made with our engine: the same stripes, panels and racket pocket in every shot, from the court to a rainy street.

You can see more product videos made this way on our samples page.

What to ask before you order AI product video

The answers to five questions tell you whether a vendor controls the product’s size or hopes for it.

Question A good answer A red flag
Where does the product’s size come from? From real dimensions, set before anything is generated “We describe it in the prompt”
How is the size checked? In every frame, with a number and a tolerance “Our team watches every video”
What happens to a frame that drifts? It is regenerated before you see the video You find it in review and ask for a revision
Will the product match across videos? Yes: one product model is reused for every video Every video is generated from scratch
Can you show one product in three videos side by side? Yes, on your own product Only single showcase videos

If you want to see the answers on your own catalog, we start with a pilot on 3⁠–⁠5 products: send a link to your catalog, and judge the videos against your own product pages. How this works for a whole catalog is on our page for marketplaces.

Sources

  1. Quelta scale test, November 5, 2026: per-frame measurements (CSV)
  2. Salsify, 2025 Consumer Research Executive Summary (survey of 1,910 U.S. and U.K. shoppers, October 2024)
  3. Amazon, Amazon Shoppable Videos: guide for sellers (US, December 2024)
  4. Amazon, How to add videos to product pages
  5. Alibaba Group, AI Powers Large-scale Applications at the World's Largest Shopping Festival (October 2025)
  6. TikTok Shop Seller University: making shoppable videos from product images

Quelta for catalogs

Need videos like this for your whole catalog?

Quelta makes product videos from your photos, adapted for each of your markets by meaning, not just by words, and checked before delivery. Start with a pilot on one category.

Start with a pilot

Tell us about your catalog

Send a link to your catalog or website and tell us roughly how many videos you need a month. We start with a pilot on 3⁠–⁠5 of your products, so you can judge the videos on your own catalog.

  1. Send a link to your catalog

    Your website, a product feed or your store on a marketplace.

  2. We make a pilot on 3⁠–⁠5 of your products

    Product-accurate videos of your real products, for the markets and platforms you need.

  3. A monthly contract for the volume you need

    New videos for the catalog and new arrivals every month, in one style.

We don’t copy other people’s videos. Every video is new and about your product. We keep the mechanics that sell (script, pacing, delivery), not anyone’s footage, people or music.

Contact us

Prefer to talk first? Yegor, Co-founder. A 30-minute video call about your catalog.

Book a call

Made with our engineTests, not client projects

Questions and answers

Why does AI change the size of my product in a video?

Generators learn sizes from the photos they were trained on. For common products, such as a tote bag, they have seen enough examples to get the scale right. For a less common product, they infer the size from the product type, and a product photo on a white background carries no scale at all. In our test, that guess came out about 1.5 times too big for a massage roller.

Will adding the dimensions to the prompt fix the size?

Not reliably. In our test, stating the real length of the roller and the height of the woman moved the median error only from 1.48× to 1.45×. Of 16 frames with the real sizes in the prompt, none came within 10% of true size.

How can I check the size in an AI product video?

Find something with a known size in the frame, such as a person, a hand or a door. Measure the product and that reference in pixels, then compare their ratio with the ratio in real life. A result within 5% of 1.00 is a pass. The calculator in this article does the math.

Does it matter if the product looks bigger than it is?

Yes. A product page that shows the wrong size sets the wrong expectation, and a product that doesn't match the page comes back. In a survey Salsify ran in October 2024, 71% of U.S. and U.K. shoppers said they had returned an item because of incorrect product content.

How does Quelta keep the product's size right?

Our engine builds a simple 3D model of each product at its real dimensions before any frame is generated. Every frame is checked against that model and the product photos, and frames that drift are regenerated before the video is assembled.